- From: David Booth <dbooth@w3.org>
- Date: Tue, 06 Jan 2004 10:45:31 -0500
- To: public-webarch-comments@w3.org
Section 1 (Introduction): Section 1 defines "Web agents" as "people or software". This is a marked departure from the definition of the term "agent" in an early Web Arch draft ( http://www.w3.org/TR/2003/WD-webarch-20030627/ ), which defines agents as "programs acting on behalf of a person, entity, or process", and a departure from common computer science usage. In my experience, in the context of software or computer science (as opposed to real estate, for example), "agent" has meant a non-human entity -- a piece of software. Furthermore, as far as I know "user agent" has always meant a piece of software. I don't think the Web Arch definition of "Web agent" or "agent" should include humans. Roughly: s/agent/agent or human/g 2.2 URI Ownership: Following the lessons of the "deep linking" debacle, it might be good to say explicitly what rights "URI ownership" does or does not confer. This is somewhat addressed later, but it might be good to say something in this section. 4.5.4 Typo (extra word): 'Nadia receives a representation data from "weather.example.com" in an unfamiliar data format.' should be 'Nadia receives a representation from "weather.example.com" in an unfamiliar data format.' -- David Booth W3C Fellow / Hewlett-Packard Telephone: +1.617.253.1273
Received on Tuesday, 6 January 2004 10:45:33 UTC